STM32F103RC board.

After soldering and cleaning operations i try to test my dc-dc converter. For this test i bye special resistor SQP5-4R7. I connected it fot output of DC-DC controller and measured current near 1 Аmp. During the test resistor heated till 110*С (measured by pyrometer). This led to falling resistance of resistor and grown up current till 1 Amp. Some photo of this tests you can see at figure 2.

Honestly this is the first time when I decided to make an industry-manufactured PCB board with MCP73833. The previous time I looked at the fact that this controller generally charges and works + became familiar with the algorithm of its operation. In this case, since the passing of the exams was approaching, and I really wanted to try it, I just sorted out the previous board and carefully transferred all the components to a new one. As a result, when I charged the battery after the previous test, I noticed that with a charging current above 375 mA, the temperature of the charge controller did not reach 45 * C. This suggests that everybody can safely solder the resistor controlling the charge in order to increase the charging current. I’ll raise it to 500mA, but I will not go further. This is not due to the heat that will be released on the chip, but to the fact that not all of my five-volt power supplies for mobile phones give out more than 500mA. Special protection, in those that I saw, no, and this can lead to problems in the electrical network.

At the end of the tests I took a beautiful photo (Figure 3), on which two LED’s are burning, indicating the end of the battery charge.

At the end of these tests, and also imagining how I will work later with this board, I found out a number of factors that I did not consider in this project:

• I did not have enough experience in PCB routing in order to install the on / off switch on the battery in this circuit board. Of course, I can do it perfectly by pulling out 18650 from the connector or by installing a special plastic plate. But this is my crutch. It would be nice to add the switch.

• There is no BMS at the battery on the board, although it is needed. Purely technically, the TPS63000 operates in the input voltage range of 1.8-6V. This is possible with a deep discharge of the battery will lead to its spoilage. This problem is solved by some modification of the battery itself and setting it cheap BMS-board directly on the battery with cramming / trimming one of the contacts. Again, not critical, but with some bugs.

• There is nothing connected in the USB connector on the digital lines. In this controller there is a built-in USB-interface, but I did not bring it to the connector. This is due to the fact that in current projects it is not necessary and to put the expellant binding of the usb interface on the board (with shielding diodes and level matching chips, what I did not have).

• There is no PowerGood LED on the board. When this device works, it does not say anything about it. This is done deliberately to extend the life of the battery, but sometimes it is not convenient. In the future I think to make a light-emitting diode on a little (or unused) pin and give a link to the LED turn-on code when the controller.

• Insufficient compactness of the board. The current dimensions of the board are 34×80 mm. I always want to get a compact solution. Maybe it will. Although on the other hand, there is not much room on the back of the PCB. What can be seen in the figure 4.

• Someone do not have a very convenient programming connector and the button of reset. Reach out to her is rather difficult, but again you can use screwdriver (crutches).

Currently there are 2 such device, so that anyone can buy it for 1,700 rubles without delivery and 18650 battery. Why don’t I send the battery? — it’s in the mail of Russia. I would be happy if they sent the batteries, although I will try to send one 18650 jar directly to the first orderer with the device for free.

You can contact me by mail: netolkomilo@ya.ru

In my opinion about the price: now for this money you can buy the original Arduino or F4Discovery from China. But neither of these you cann’t take with you for a whole day without an external kit and batteries. And then you get a finished piece, in which you can insert your board and get a very rigid design, which you can safely carry with you without worrying that somewhere the soldering will disappear or the device will experience a short circuit.

Well, lastly pinout the findings.

So it turned out that preparing a fee for the first sale (not my above-sampled sample), I decided to finish in it a number of features, for the consumer. As a result, the following was obtained:

You may notice the following changes:

• There is no «comb» PBS-30 for insertion into the printed circuit board. To my regret, there are 2 factors working here (they ran out and the customer asked them not to put it.)

• The programming connector is made with other angled terminals. Last time, I also wanted to use such connector, but they were not in the available store. As a result, I accidentally skidded to one of the Moscow radio markets, where I found the widest choice of these connectors and bought this ones.

• LEDs are slightly different. To my regret, I often take SMD-LEDs from a large heap, and I don’t know what color they are. This time it’s two green and one white LED.

But the devil is often covered in detail, so it turns out here. On this board, I tried many different resistors in controlling the charging current. As a result, it was revealed that the MCP73833 on this board normally holds a current of 800mA and when it is charged with 2.5V Li-Ion batteries it warms up to 65-70 * C. Of course, it works with currents up to 2A, but I honestly do not know how to implement them on the msop-10 case. Maybe in the next building, in which this problem is solved better, it will be possible to extend 2A.

An additional small detail, which few people will notice at all, is the installation of inductance from Murata and not its Chinese twin brother with an unknown name. They look like the same, but this one is better screened.

The general view of micro-changes is seen in Figure 8.

During the full diagnostic test I destroy one STM32F103RCT6. How I did it:

I took an ordinary multimeter and try to measure voltage into a feedback resistor. This TPS63000 sees a voltage dropout on the feedback and begins to rapidly increase the voltage at its output. As a result, according to the test with the sealed controller: the voltage in the pulse grew to 7V. In this case, the controller undergoes an overvoltage and dies.

Now i try to do some projects which use this board. I think that i write something to describe my projects.